From 4e83de107909467c44147eb55c69f007570b3f5f Mon Sep 17 00:00:00 2001 From: senseb Date: Mon, 6 Jan 2025 19:48:29 +0800 Subject: [PATCH 1/3] mark deprecated models in siliconflow #12399 --- .../model_providers/siliconflow/llm/_position.yaml | 3 --- .../model_providers/siliconflow/llm/hunyuan-a52b-instruct.yaml | 1 + .../model_providers/siliconflow/llm/internvl2-llama3-76b.yaml | 1 + .../siliconflow/llm/qwen2.5-math-72b-instruct.yaml | 1 + 4 files changed, 3 insertions(+), 3 deletions(-) diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/_position.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/_position.yaml index 8703a97edd1133..a4d1aac10cb46a 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/_position.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/_position.yaml @@ -1,4 +1,3 @@ -- Tencent/Hunyuan-A52B-Instruct - Qwen/QwQ-32B-Preview - Qwen/Qwen2.5-72B-Instruct - Qwen/Qwen2.5-32B-Instruct @@ -6,11 +5,9 @@ - Qwen/Qwen2.5-7B-Instruct - Qwen/Qwen2.5-Coder-32B-Instruct - Qwen/Qwen2.5-Coder-7B-Instruct -- Qwen/Qwen2.5-Math-72B-Instruct - Qwen/Qwen2-VL-72B-Instruct - Qwen/Qwen2-1.5B-Instruct - Pro/Qwen/Qwen2-VL-7B-Instruct -- OpenGVLab/InternVL2-Llama3-76B - OpenGVLab/InternVL2-26B - Pro/OpenGVLab/InternVL2-8B - deepseek-ai/DeepSeek-V2.5 diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/hunyuan-a52b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/hunyuan-a52b-instruct.yaml index c5489554a60456..51d6c024ff5325 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/hunyuan-a52b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/hunyuan-a52b-instruct.yaml @@ -82,3 +82,4 @@ pricing: output: '21' unit: '0.000001' currency: RMB +deprecated: true diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/internvl2-llama3-76b.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/internvl2-llama3-76b.yaml index 65386d3175fa1c..b5443df18c0f8c 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/internvl2-llama3-76b.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/internvl2-llama3-76b.yaml @@ -82,3 +82,4 @@ pricing: output: '21' unit: '0.000001' currency: RMB +deprecated: true diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-math-72b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-math-72b-instruct.yaml index 1b6f2603f5eabe..40c9ab48ca1e63 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-math-72b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2.5-math-72b-instruct.yaml @@ -82,3 +82,4 @@ pricing: output: '4.13' unit: '0.000001' currency: RMB +deprecated: true From fa4507729860a99624933de4eec9aad160a30960 Mon Sep 17 00:00:00 2001 From: senseb Date: Tue, 7 Jan 2025 06:57:55 +0800 Subject: [PATCH 2/3] Fixes #12407 --- .../siliconflow/llm/qwen-qvq-72B-preview.yaml | 54 +++++++++++++++++++ .../siliconflow/llm/qwen-qwq-32B-preview.yaml | 4 +- .../llm/qwen2-vl-72b-instruct.yaml | 4 +- .../siliconflow/llm/qwen2-vl-7b-Instruct.yaml | 4 +- 4 files changed, 60 insertions(+), 6 deletions(-) create mode 100644 api/core/model_runtime/model_providers/siliconflow/llm/qwen-qvq-72B-preview.yaml diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qvq-72B-preview.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qvq-72B-preview.yaml new file mode 100644 index 00000000000000..dada6bb803f0b2 --- /dev/null +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qvq-72B-preview.yaml @@ -0,0 +1,54 @@ +model: Qwen/QVQ-72B-Preview +label: + en_US: Qwen/QVQ-72B-Preview +model_type: llm +features: + - agent-thought + - tool-call + - stream-tool-call + - vision +model_properties: + mode: chat + context_size: 32768 +parameter_rules: + - name: temperature + use_template: temperature + - name: max_tokens + use_template: max_tokens + type: int + default: 8192 + min: 1 + max: 16384 + help: + zh_Hans: 指定生成结果长度的上限。如果生成结果截断，可以调大该参数。 + en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. + - name: top_p + use_template: top_p + - name: top_k + label: + zh_Hans: 取样数量 + en_US: Top k + type: int + help: + zh_Hans: 仅从每个后续标记的前 K 个选项中采样。 + en_US: Only sample from the top K options for each subsequent token. + required: false + - name: frequency_penalty + use_template: frequency_penalty + - name: response_format + label: + zh_Hans: 回复格式 + en_US: Response Format + type: string + help: + zh_Hans: 指定模型必须输出的格式 + en_US: specifying the format that the model must output + required: false + options: + - text + - json_object +pricing: + input: '9.90' + output: '9.90' + unit: '0.000001' + currency: RMB diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qwq-32B-preview.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qwq-32B-preview.yaml index c949de4d75604c..e73c5d2030d458 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qwq-32B-preview.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen-qwq-32B-preview.yaml @@ -15,9 +15,9 @@ parameter_rules: - name: max_tokens use_template: max_tokens type: int - default: 512 + default: 4096 min: 1 - max: 4096 + max: 8192 help: zh_Hans: 指定生成结果长度的上限。如果生成结果截断，可以调大该参数。 en_US: Specifies the upper limit on the length of generated results. If the generated results are truncated, you can increase this parameter. diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-72b-instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-72b-instruct.yaml index 1866a684b55b60..f5180b41f1ea50 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-72b-instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-72b-instruct.yaml @@ -78,7 +78,7 @@ parameter_rules: - text - json_object pricing: - input: '21' - output: '21' + input: '4.13' + output: '4.13' unit: '0.000001' currency: RMB diff --git a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-7b-Instruct.yaml b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-7b-Instruct.yaml index a50834468922f9..0ffbaee3837991 100644 --- a/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-7b-Instruct.yaml +++ b/api/core/model_runtime/model_providers/siliconflow/llm/qwen2-vl-7b-Instruct.yaml @@ -78,7 +78,7 @@ parameter_rules: - text - json_object pricing: - input: '21' - output: '21' + input: '0.35' + output: '0.35' unit: '0.000001' currency: RMB From c20ad195edb017db67af433800ea47ab856129dd Mon Sep 17 00:00:00 2001 From: crazywoola <427733928@qq.com> Date: Tue, 7 Jan 2025 09:04:56 +0800 Subject: [PATCH 3/3] fix: lint --- docker/docker-compose-template.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docker/docker-compose-template.yaml b/docker/docker-compose-template.yaml index baea9844f4808a..9ae7718d7f255d 100644 --- a/docker/docker-compose-template.yaml +++ b/docker/docker-compose-template.yaml @@ -1,4 +1,4 @@ -x-shared-env: &shared-api-worker-env +x-shared-env: &shared-api-worker-env services: # API service api: